{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:R5GAB3QUN4ROPZELBQKGXCHRNY","short_pith_number":"pith:R5GAB3QU","canonical_record":{"source":{"id":"2507.17797","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-23T15:22:51Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"70d2b6e25e8cd557396e5dada8c782fa75a8ce507243d492b147d558635b9f9c","abstract_canon_sha256":"5af1dd95aa8d59df0cf312ec16d56c82526b9274ad6245547d1e1ad2f6937056"},"schema_version":"1.0"},"canonical_sha256":"8f4c00ee146f22e7e48b0c146b88f16e3c8cc38fef6d043599182f03f7f061e4","source":{"kind":"arxiv","id":"2507.17797","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.17797","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"arxiv_version","alias_value":"2507.17797v1","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17797","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"pith_short_12","alias_value":"R5GAB3QUN4RO","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"pith_short_16","alias_value":"R5GAB3QUN4ROPZEL","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"pith_short_8","alias_value":"R5GAB3QU","created_at":"2026-07-05T11:42:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:R5GAB3QUN4ROPZELBQKGXCHRNY","target":"record","payload":{"canonical_record":{"source":{"id":"2507.17797","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-23T15:22:51Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"70d2b6e25e8cd557396e5dada8c782fa75a8ce507243d492b147d558635b9f9c","abstract_canon_sha256":"5af1dd95aa8d59df0cf312ec16d56c82526b9274ad6245547d1e1ad2f6937056"},"schema_version":"1.0"},"canonical_sha256":"8f4c00ee146f22e7e48b0c146b88f16e3c8cc38fef6d043599182f03f7f061e4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:35.447562Z","signature_b64":"EYZSI7fglkGg/YssIbTWZsuaQySqPl8z9CpN76/xlhUhSVae0H8zaROIgJisp4P+M+xDjFp1TyT694ET00/CDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f4c00ee146f22e7e48b0c146b88f16e3c8cc38fef6d043599182f03f7f061e4","last_reissued_at":"2026-07-05T11:42:35.446966Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:35.446966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.17797","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:42:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BRXg6De371K/j4MRbpgnJIpH9atNPciw527Ox+JvcVPqtIXnteGEo5cQGyoIUKpPXQhRIfO3H5QpMOTdi3TBAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:27:11.410923Z"},"content_sha256":"702ba9af638b564574ba40397776630e07150646920dade54a4b58b81c6d3e8e","schema_version":"1.0","event_id":"sha256:702ba9af638b564574ba40397776630e07150646920dade54a4b58b81c6d3e8e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:R5GAB3QUN4ROPZELBQKGXCHRNY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GenSelect: A Generative Approach to Best-of-N","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Aleksander Ficek, Igor Gitman, Ivan Moshkov, Ivan Sorokin, Shubham Toshniwal","submitted_at":"2025-07-23T15:22:51Z","abstract_excerpt":"Generative reward models with parallel sampling have enabled effective test-time scaling for reasoning tasks. Current approaches employ pointwise scoring of individual solutions or pairwise comparisons. However, pointwise methods underutilize LLMs' comparative abilities, while pairwise methods scale inefficiently with larger sampling budgets. We introduce GenSelect, where the LLM uses long reasoning to select the best solution among N candidates. This leverages LLMs' comparative strengths while scaling efficiently across parallel sampling budgets. For math reasoning, we demonstrate that reason"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17797","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2507.17797/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:42:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9kjpOGCieO+Er8n59yuSAaARlW0CSyyQXlTxkB3P08FN+OiSq2hIt3FDHD1ikdSSFAFhgsow53a3GzM29wERBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:27:11.411494Z"},"content_sha256":"aac2841703eb17876a952e748f0186de0781c675d47ef8d47bba20cc4eacaf84","schema_version":"1.0","event_id":"sha256:aac2841703eb17876a952e748f0186de0781c675d47ef8d47bba20cc4eacaf84"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R5GAB3QUN4ROPZELBQKGXCHRNY/bundle.json","state_url":"https://pith.science/pith/R5GAB3QUN4ROPZELBQKGXCHRNY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R5GAB3QUN4ROPZELBQKGXCHRNY/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T05:27:11Z","links":{"resolver":"https://pith.science/pith/R5GAB3QUN4ROPZELBQKGXCHRNY","bundle":"https://pith.science/pith/R5GAB3QUN4ROPZELBQKGXCHRNY/bundle.json","state":"https://pith.science/pith/R5GAB3QUN4ROPZELBQKGXCHRNY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R5GAB3QUN4ROPZELBQKGXCHRNY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:R5GAB3QUN4ROPZELBQKGXCHRNY","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"5af1dd95aa8d59df0cf312ec16d56c82526b9274ad6245547d1e1ad2f6937056","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-23T15:22:51Z","title_canon_sha256":"70d2b6e25e8cd557396e5dada8c782fa75a8ce507243d492b147d558635b9f9c"},"schema_version":"1.0","source":{"id":"2507.17797","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.17797","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"arxiv_version","alias_value":"2507.17797v1","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17797","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"pith_short_12","alias_value":"R5GAB3QUN4RO","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"pith_short_16","alias_value":"R5GAB3QUN4ROPZEL","created_at":"2026-07-05T11:42:35Z"},{"alias_kind":"pith_short_8","alias_value":"R5GAB3QU","created_at":"2026-07-05T11:42:35Z"}],"graph_snapshots":[{"event_id":"sha256:aac2841703eb17876a952e748f0186de0781c675d47ef8d47bba20cc4eacaf84","target":"graph","created_at":"2026-07-05T11:42:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2507.17797/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Generative reward models with parallel sampling have enabled effective test-time scaling for reasoning tasks. Current approaches employ pointwise scoring of individual solutions or pairwise comparisons. However, pointwise methods underutilize LLMs' comparative abilities, while pairwise methods scale inefficiently with larger sampling budgets. We introduce GenSelect, where the LLM uses long reasoning to select the best solution among N candidates. This leverages LLMs' comparative strengths while scaling efficiently across parallel sampling budgets. For math reasoning, we demonstrate that reason","authors_text":"Aleksander Ficek, Igor Gitman, Ivan Moshkov, Ivan Sorokin, Shubham Toshniwal","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-23T15:22:51Z","title":"GenSelect: A Generative Approach to Best-of-N"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17797","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:702ba9af638b564574ba40397776630e07150646920dade54a4b58b81c6d3e8e","target":"record","created_at":"2026-07-05T11:42:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"5af1dd95aa8d59df0cf312ec16d56c82526b9274ad6245547d1e1ad2f6937056","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-23T15:22:51Z","title_canon_sha256":"70d2b6e25e8cd557396e5dada8c782fa75a8ce507243d492b147d558635b9f9c"},"schema_version":"1.0","source":{"id":"2507.17797","kind":"arxiv","version":1}},"canonical_sha256":"8f4c00ee146f22e7e48b0c146b88f16e3c8cc38fef6d043599182f03f7f061e4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8f4c00ee146f22e7e48b0c146b88f16e3c8cc38fef6d043599182f03f7f061e4","first_computed_at":"2026-07-05T11:42:35.446966Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:35.446966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EYZSI7fglkGg/YssIbTWZsuaQySqPl8z9CpN76/xlhUhSVae0H8zaROIgJisp4P+M+xDjFp1TyT694ET00/CDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:35.447562Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.17797","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:702ba9af638b564574ba40397776630e07150646920dade54a4b58b81c6d3e8e","sha256:aac2841703eb17876a952e748f0186de0781c675d47ef8d47bba20cc4eacaf84"],"state_sha256":"178e102e7551bf9b47cdcef1989a3139f74c48850b5737259096ff15c837ba52"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VYFmZ2HqZfTgJR1fK7BrBdyo6cTrSBjeRGe8yo2S3WcgMcN5HBZaGEtpUCmrqkFqzh067WOK9PN93G3ALvCzCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T05:27:11.416161Z","bundle_sha256":"3fcfc2e08a583f4fea0a64cdb14099a159b2df1134a3826cd6b73809626cbe0b"}}